The invention relates to an urban traffic flow prediction method based on road network spatial heterogeneity. The method comprises the following steps: obtaining a traffic topological structure and a historical traffic flow data set, and constructing a traffic characteristic matrix, an adjacent matrix and a mapping function; a spatial heterogeneity detection module is adopted to carry out flow level data enhancement and graph topology level structure enhancement on the traffic characteristic matrix, the adjacent matrix and the mapping function, then a neural network model is built and trained, and the trained neural network model is used to carry out traffic flow prediction. According to the invention, prediction is carried out by using the time-space correlation information of the urban road traffic flow, and the defect that the existing method cannot fully utilize the time-space characteristics and periodic characteristics of the traffic flow data can be overcome; and meanwhile, different traffic modes of different road sections are combined, and the spatial features are subjected to graph enhancement, so that the accuracy of traffic flow prediction is improved.
本申请涉及一种基于路网空间异质性的城市交通流预测方法,它包括如下步骤:获取交通拓扑结构和历史交通流数据集并构建交通特征矩阵、邻接矩阵和映射函数;采用空间异质性检测模块对交通特征矩阵、邻接矩阵和映射函数进行流量级数据增强和图拓扑级结构增强后搭建神经网络模型并进行训练,使用训练好的神经网络模型进行交通流量预测。本发明利用城市道路交通流的时空关联信息进行预测,能够克服现有方法无法充分利用交通流数据的时空特征和周期性特征的缺点;同时将不同路段的不同交通模式结合起来,并对空间特征进行图增强,提高交通流预测的准确性。
Urban traffic flow prediction method based on road network spatial heterogeneity
一种基于路网空间异质性的城市交通流预测方法
20.09.2024
Patent
Elektronische Ressource
Chinesisch
Traffic flow prediction method and system for urban road network
Europäisches Patentamt | 2021
|Fine-grained urban traffic flow inference method based on spatial heterogeneity
Europäisches Patentamt | 2024
|Urban traffic road network traffic flow prediction method considering carbon emission model
Europäisches Patentamt | 2024
|Road network flow prediction method for urban traffic signal control
Europäisches Patentamt | 2023
|